Weed Detection in Peanut Fields Based on Machine Vision

Abstract

The accurate identification of weeds in peanut fields can significantly reduce the use of herbicides in the weed control process. To address the identification difficulties caused by the cross-growth of peanuts and weeds and by the variety of weed species, this paper proposes a weed identification model named EM-YOLOv4-Tiny incorporating multiscale detection and attention mechanisms based on YOLOv4-Tiny. Firstly, an Efficient Channel Attention (ECA) module is added to the Feature Pyramid Network (FPN) of YOLOv4-Tiny to improve the recognition of small target weeds by using the detailed information of shallow features. Secondly, the soft Non-Maximum Suppression (soft-NMS) is used in the output prediction layer to filter the best prediction frames to avoid the problem of missed weed detection caused by overlapping anchor frames. Finally, the Complete Intersection over Union (CIoU) loss is used to replace the original Intersection over Union (IoU) loss so that the model can reach the convergence state faster. The experimental results show that the EM-YOLOv4-Tiny network is 28.7 M in size and takes 10.4 ms to detect a single image, which meets the requirement of real-time weed detection. Meanwhile, the mAP on the test dataset reached 94.54%, which is 6.83%, 4.78%, 6.76%, 4.84%, and 9.64% higher compared with YOLOv4-Tiny, YOLOv4, YOLOv5s, Swin-Transformer, and Faster-RCNN, respectively. The method has much reference value for solving the problem of fast and accurate weed identification in peanut fields.

An Improved U-Net Model Based on Multi-Scale Input and Attention Mechanism: Application for Recognition of Chinese Cabbage and Weed

“…Additionally, as can be seen in Figure 7a, the improved MSECA-Unet model converged after 130 iterations, stabilized near the highest value earlier, and did so significantly more quickly than the other three comparison models. This is because, in this paper, the ECA module, which can successfully prevent the activation of irrelevant information and noise in the network, is introduced before the fusion of features in the U-Net network, so that it only fuses the feature information that requires attention, which decreases the time loss in feature fusion, and hence, quickens the model's convergence, which is consistent with the conclusions reached by Zhang et al [29] when introducing the ECA module into the YOLOv4-Tiny network, and by Zhao et al [52] when introducing the ECA module into DenseNet network.…”

“…Target detection can be considered as a combination of two tasks, target localization and classification, i.e., locating the position of an object in an image and identifying the class to which the object belongs [28]. To quickly and accurately recognize different types of weeds in peanut fields, Zhang et al [29] proposed the EM-YOLOv4-Tiny weed identification model based on the YOLOv4-Tiny target detection model, multiscale detection, and the attention mechanism; Kang et al [30] proposed a weed detection method in sugar beet based on SSD model based on multi-scale fusion module and feature improvement; Partel et al [31] proposed three improved target detection models based on the YOLOv3 model and implemented a precision intelligent sprayer for weed recognition with sunflower and weed and pepper and weed as datasets; Peng et al [32] compared two target detection networks, Faster R-CNN and YOLOv3, and structurally optimized the Faster R-CNN network by introducing a feature pyramid network in the RPN network to generate target candidate frames to achieve the efficient identification of cotton field weeds in complex backgrounds. The first semantic segmentation model, fully convolutional networks (FCN), which segmented images by end-to-end training of convolutional neural networks, was suggested by Long et al in 2015 [33].…”

Weed target detection at seedling stage in paddy fields based on YOLOX

“…The recall (R) is defined as the ratio of the number of positive classes predicted to be positive to the number of all real positive classes. Its mathematical expression is shown in Eq (6).…”

“…With the continuous development of modern agricultural production technology, machine vision technology and deep learning have been applied for weed target detection in precision agriculture [ 6 , 7 ]. Traditional weed spraying systems spray pesticides regardless of the presence of weeds.…”

Evolution of Agriculture : A Survey

“…Soil moisture content, land surface temperature, and the calculation of the soil moisture index may all be sensed using optical remote sensing, synthetic aperture radar, and thermal remote sensing technologies. Precision irrigation water management at a big scale is the consequence of this data analysis.In [44],author's have concluded that, one of the most important oil crops in the world and a key component of the world's oil production is the peanut. To quickly and precisely identify the many weed species that can be found in peanut fields.…”